Recently I’ve been thinking that it might be possible to test a subset of creationist arguments: that molecular similarities are better explained by common design than common descent.
I believe that using a Bayes factor-based analysis along the lines of Theobald (2010) would be the appropriate approach. Not only would this force the construction a positive model of design (a first in creationism), but one can also integrate over the different intelligent designers that the DI likes to throw out there. So whether Hanuman, God, Jesus, God, the Holy Ghost, God, Richard Dawkins, God, Moses, God, Uranus, God, Thor, God, Prof. X, or God is the true common designer (or a mixture of the ones above), they all can be included in the model.
Of course, ID advocates would probably insist on using a Dirac delta function as the prior.